A review of greenhouse gas emission factors for fertiliser production.

A Review of Greenhouse Gas Emission Factorsfor Fertiliser Production.Sam Wood and Annette CowieResearch and Development Division, State Forests of New South Wales.Cooperative Research Centre for Greenhouse AccountingFor IEA Bioenergy Task 38June 2004

1. IntroductionGiven their significant contribution to rising atmospheric greenhouse gasconcentrations, accounting for emissions of CO 2 , N 2 O and CH 4 from agriculturalpractices has become increasingly important. Emissions of these gases may occureither directly during agricultural activities (eg. cultivation and harvesting), orindirectly during the production and transport of required inputs (eg. herbicides,pesticides and fertilisers). Life Cycle Assessment (LCA) is commonly employed toundertake a complete evaluation of emissions. In LCA the environmental impacts ofproducts and processes are analysed from ‘cradle to grave’, such that both direct andindirect emissions from agricultural practices are included.The production of fertilisers demands much energy and generates considerablegreenhouse gas (GHG) emissions. Kongshaug (1998) estimates that fertiliserproduction consumes approximately 1.2% of the world’s energy and is responsible forapproximately 1.2% of the total GHG emissions. As such, fertiliser productioncomprises an important component of agricultural Life Cycle Assessments wheresystem boundaries are wide enough to include indirect emissions from agriculturalinputs. Directly calculating GHG emissions from fertiliser production for individualLCA studies is problematic given the large variety of fertilisers used and the complexprocesses involved in their production. Furthermore, the emissions data required forsuch calculations are either difficult to obtain or constrained in their extent and quality(Patyk 1996).Emission factors provide a useful shortcut for use in LCA, avoiding the need fordetailed calculations of emissions. An emission factor is a typical quantity of GHGsreleased to the atmosphere per unit of activity, in this case, per unit weight of fertiliserproduced (i.e. g CO 2-e / kg fertiliser). Since fertiliser emission factors vary widelydepending on production technology, it is preferable to use customised emissionfactors relevant to the particular plant from which the fertiliser under considerationwas produced. Unfortunately, such information is rarely available. The objective ofthis report is to collate published emission factors for GHGs associated with theproduction of a range of nitrogen, phosphate and multi-nutrient fertilisers, for use as2

inputs to agricultural and forestry Life Cycle Assessments and calculations ofgreenhouse gas balance.1.1. Scope.During the life cycle of fertiliser products, GHG emissions may arise during theextraction of resources, the transport of raw materials and products, and duringfertiliser production processes. Where possible, separate emissions factors areprovided for each step in this life cycle. Emissions arising from the application offertilisers in the field were not considered. A brief discussion of key processesinvolved in the production of each type of fertiliser, and their associated emissions, isincluded in this report.Emissions factors for similar fertiliser products differ markedly between reports,depending on certain aspects of plant design and efficiency, emissions controlmechanisms and raw material inputs. The assumptions made by the analysts duringthe calculations of emission factors also contribute to this variation. Ideally, detailedinformation regarding the cause of between-source variation would be provided hereto guide the selection of the appropriate emission factor for particular LCA’s.However, the extent to which this type of information was documented in thereviewed studies was often limited (see ‘Transparency’ in Table 1). As a result of thislack of transparency, comparisons of emission factors were considered beyond thescope of this review, and only limited discussions of between-source variability couldbe provided.GHG emissions from fertiliser production are closely associated with energyconsumption. Most studies reviewed here coupled emissions estimates with detailedenergy budgets (see Table 1). Data on energy consumption during fertilisermanufacture are not presented here, but discussion of the relationship betweenemissions factors and energy consumption are included in the text.3

1.2. UnitsThe significant GHG emissions arising from the production of fertilisers are carbondioxide (CO 2 ), nitrous oxide (N 2 O) and methane (CH 4 ). In line with internationalgreenhouse accounting practice (IPCC 1996a), emission factors are expressed ascarbon dioxide equivalents per unit mass of fertiliser product (eg. g CO 2-e / kgfertiliser) or element (eg. g CO 2-e / kg N). This involved some simple calculations:i. CO 2 , N 2 O and CH 4 emissions were converted to CO 2 equivalents (CO 2-e ) usingthe ‘global warming potential’ (GWP), which determines the relativecontribution of a gas to the greenhouse effect. The GWP (with a time span of100 years) of CO 2 , CH 4 and N 2 O is 1, 21 and 310, respectively (IPCC 1996a).CO 2-e emissions for each gas (CO 2 , N 2 O and CH 4 ) were summed together to giveTotal CO 2-e . A CO 2 :N 2 O:CH 4 ratio is included to convey the relativecontribution of each gas to Total CO 2-e .ii. Conversion of mass/product to mass/element (or vice-versa) was based on thestated composition of the fertiliser product (i.e. NPK fertiliser: 24% N, 7% P and3% K). If the composition of the product was unavailable, the ratio was assumedto correspond with equivalent products from other studies.4

2. Summary of Literature.There are few published studies that present emission factors for GHGs arising fromfertiliser production (Table 1). All but one of these studies was conducted in WesternEurope and analyses were generally coupled with energy LCA. It is likely that moreemissions data exist within unpublished reports held within the fertiliser industry or ascomponents of the multitude of agricultural LCA databases available. The studies inTable 1 differed in their consideration of certain components in the life cycle offertiliser manufacturing, particularly in relation to transport and GHGs other thanCO 2 .Table 1: Summary of published studies into GHG emissions from Fertiliser Production.Reference Country Type Results Transparency bEnergy CO 2 N 2O CH 4 Transportto manuf.Transportto fieldsDavis andSweden and Masters Yes Yes Yes Yes Yes No YesHaglund (1999) WesternEuropeThesisKongshaug Western Conference Yes Yes Yes No No No Partial(1998)Europe ProceedingsKuesters and Europe Conference Yes Yes Yes Yes Yes Yes NoJenssen (1998)proceedingsKramer et al. Netherlands Journal No Yes Yes Yes Yes No Partial(1999)articleElsayed et al. Europe Report Yes Yes Yes Yes ? ? Partial(2003)Patyk (1996) Germany Conference Yes Yes Yes Yes Yes Yes PartialproceedingsPatyk andGermany Conference Yes Yes Yes Yes Yes Yes PartialReinhardt (1996)proceedingsKaltschmitt andReinhardt (1997) a Germany Book Yes Yes Yes Yes ? ? ?West and Marland(2001)USAJournalarticleYes Yes No No Yes Yes Partiala) Kaltschmitt and Reinhardt (1997), cited in Elsayed et al. (2003).b) Transparency refers to the detail provided on the calculation of emissions factors for fertiliser production.Note: Studies citing emission factors for intermediate products such as Ammonia and Nitric Acid were not included in this table.3. Greenhouse Gas Emission Factors for Nitrogen Fertiliser Production.In this section, published GHG emission factors for key nitrogen fertilisers arepresented and discussed. Emission factors for Ammonium Nitrate (AN), CalciumAmmonium Nitrate (CAN), Urea, Urea Ammonium Nitrate (UAN) and a ‘MeanNitrogen Fertiliser’ were available. Given their importance as the primaryintermediate products used in nitrogen fertiliser manufacturing and their importantcontribution to total emissions, emissions factors for Ammonia and Nitric Acidproduction are also presented.5

1998) or be considered to replace the combustion of fossil fuels elsewhere in the lifecycle (eg. Davis and Haglund 1999). Table 2 provides a summary of how somestudies interpreted this energy and emissions credit within the fertiliser life cycle.Table 2: Interpretation of emissions credits from steam exports during fertiliser production.ReferenceSteamCreditNotes from reference regarding emissions credit for steamproductionDavis and Haglund (1999) Yes “In the case of generation of steam owing to fertiliser production,steam has been assumed to replace combustion of oil” (pp. 97).“The reactions involved in producing nitric acid reaction yield heat,which is partly transported to a district-heating network. The districtheat energy is deducted from net energy consumption” (pp. 64).Kongshaug (1998) Yes “Full credit has been given to steam export from production units. Themost efficient ammonia, nitric acid and sulphuric acid plants havetoday very high surplus of energy, especially the sulphuric acid plants”(pp. 17).West and Marland (2001) No “The energy balance credit comprised less than 2% of the total energyinput into the production of N fertiliser and was not included in theestimate of CO 2 emissions” (pp. 6).Mortimer (pers. comm.) Yes “There are many complications involved in the calculations, chiefly dueto the need to assess ammonia and nitric acid production and the needto address a substantial steam export ‘credit’ (assumed to displacesome natural gas use)”.Elsayed et al. (2003) Yes No details given.Patyk (1996) ? No discussion of steam credits was provided in the text. However, anegative emissions value for CO 2 was included in the emissionsestimates for nitrogen fertilisers, which may infer a steam credit.Patyk and Reinhardt (1996)Kuesters and Jenssesn(1998)Kramer (1999)Kaltschmitt and Reinhardt(1997)Unknown3.1.2. Emission Factors for Ammonia ProductionInsufficient information was provided to determine whether steamcredits were included in analyses.Emission factors for ammonia production are presented in Table 3. CO 2 emissionsarising from fossil fuels consumed as an energy source and feedstock dominated GHGemissions from ammonia synthesis, with emissions from other sources (eg. transport)making only minor contributions. Estimates were generally based on the quantity ofenergy consumed and the appropriate emission factors for the respective fossil fuelinputs. Natural gas was the primary fossil fuel for all estimates given in Table 3.Given the lack of transparency of several reports, it is difficult to discuss the source ofdifferences amongst emission factors. Variation can be attributed to the overallefficiency of each plant (related to plant age and design), the use of alternative fossilfuel inputs (i.e. coal and/or oil) and, to a lesser extent, the way steam exports wereinterpreted.7

Table 3: Greenhouse gas emission factors for Ammonia Production.Product Country Composition a g CO 2-e ReferenceN:P:K per kg N per kg CO 2:N 2O:CH 4ProductAmmonia Norway 82:0:0 1829.3 1500.0 - IPCC (1996a and1996b)Ammonia Netherlands 82:0:0 2637.8 2163.0 99.9:0.0:0.1 Kramer (1999)Ammonia Europe 82:0:0 2087.0 1711.3 95.6:0.1:4.0 Mortimer (pers. comm.)Ammonia Europe Average 82:0:0 2329.3 1910.0 - Kongshaug (1998)Ammonia Europe Modern 82:0:0 2024.4 1660.0 - Kongshaug (1998)Tech.Ammonia West Europe 82:0:0 1402.4- 1150-1300 - EFMA (2000a)1585.4Ammonia Canada 82:0:0 1951.2 1600.0 - IPCC (1996a and1996b)Ammonia USA (ammonia 82:0:0 1536.6 1260.0 - EPA (1993)plant)Ammonia USA 82:0:0 1491.5 1223.0 - DOE (2000)Ammonia Australia (integrated 82:0:0 1524.4- 1250-1800 - EFMA (2000h)ammonia/urea plant)2195.1a. The percentage N (by weight) of ammonia is approximately 82% (Engelstad et al. 1985; Kongshaug 1998)Note: Figures in italics are derived values, based on % N composition.3.2. Greenhouse Gas Emission Factors for Nitric Acid Production.3.2.1. Overview of Nitric Acid Production.Nitric acid is used in the manufacturing of Ammonium Nitrate, Calcium Nitrate andPotassium Nitrate, which, in turn, are used either as straight fertilisers or mixed intocompound fertilisers. Most nitric acid is produced by catalytic oxidation of ammoniaat high-pressure and high temperature. All plants producing nitric acid are based onthe same basic chemical reactions: oxidation of ammonia with air to give nitric oxide;and, oxidation of the nitric oxide to nitrogen dioxide and absorption in water to give asolution of nitric acid (EFMA 2000b). The oxidation of ammonia also yields other byproductoxides of nitrogen which are emitted as a tail gas, namely, nitrous oxide(N 2 O), nitrogen monoxide (NO, nitric oxide) and nitrogen dioxide (NO 2 ) (EFMA2000b). The reaction from ammonia to nitric acid is exothermic (heat releasing) andcontributes to a considerable net steam export, which may be considered an energyand emissions ‘credit’ in GHG accounting (see Table 2).Nitrous Oxide (N 2 O) is the most significant GHG associated with the production ofnitric acid. N 2 O is a highly ‘effective’ greenhouse gas, with a global warmingpotential 310 times greater than CO 2 (IPCC 1996a). The IPCC currently believe thatnitric acid production is the largest industrial source of N 2 O (IPCC 2000). The amountof N 2 O emitted is dependent upon combustion conditions (pressure, temperatures),catalyst composition, burner design (EFMA 2000b) and emissions abatement8

technologies (IPCC 2000). Non-Selective Catalytic Reduction (NSCR), a typical tailgas treatment in the USA and Canada, may reduce N 2 O emissions by 80-90% (IPCC2000) and a nitric acid manufacturer in Norway has developed a N 2 O abatementprocess giving 70-85% N 2 O reduction (Kongshaug 1998). Despite their advantages,an estimated 80% of the nitric acid plants worldwide do not employ NSCRtechnology (IPCC 2000).3.2.2. Emission Factors for Nitric Acid ProductionEmission factors for nitric acid production are given in Table 4. With the exception ofKramer (1999) and Mortimer (pers. comm.) the only GHG included is N 2 O. Estimatesare generally based on point source measurement of N 2 O concentration in tail gases atnitric acid plants. In LCA studies where emissions credits from steam generatedduring nitric acid production were considered (see Table 2), the credit was generallyapplied elsewhere in the fertiliser production life cycle. Mortimer (pers. comm.)presented a small emission credit in this step of the fertiliser life cycle.N 2 O emissions from nitric acid production are highly variable with estimates rangingfrom 550-2945 g CO 2-e /kg Nitric Acid (Table 4). Variation among estimates can beattributed to the installation of emissions abatement technologies. It should be notedthat emission factors as high as 5890 CO 2-e /kg Nitric Acid are given for plants notequipped with NSCR technology (IPCC 2000).Table 4: Greenhouse gas emission factors for Nitric Acid ProductionProduct Country Composition a g CO 2-e ReferenceN:P:K per kg N per kg CO 2:N 2O:CH 4ProductNitric Acid USA 22.2:0:0 2818.2- 620-2790 0:100:0 IPCC (2000)12681.8Nitric Acid Norway – modern, integrated 22.2:0:0 2818.2

Nitric Acid Europe Average 22.2:0:0 9000.0 1980.0 0:100:0 Kongshaug (1998)Nitric Acid Europe Modern Technology 22.2:0:0 2500.0 550.0 0:100:0 Kongshaug (1998)(1999)Nitric Acid Netherlands 22.2:0:0 10851.9 2387.4 22.5:77.1:0.3 Kramer (1999)Nitric Acid Europe 22.2:0:0 9035.5 1987.8 0:100:0 b Mortimer (pers. comm.)(a) From Patyk (1996). This may vary for different nitric acid producers.(b) Mortimer gave a net credit for CO 2 and CH 4 of 85 and 4 g CO 2-e/kg Nitric Acid.(c) NSCR: Non-Selective Catalytic Reduction technology, see text.Note: Figures in italics are derived values, based on % N composition given in Patyk (1996).3.3. Greenhouse Gas Emission Factors for Ammonium Nitrate, CalciumAmmonium Nitrate and Other Nitrogen Fertilisers3.3.1. Overview of AN, CAN and N Fertiliser ProductionEmissions data were available for Ammonium Nitrate (AN), Calcium AmmoniumNitrate (CAN) and ‘Mean Nitrogen Fertiliser’. AN is used extensively as anitrogenous fertiliser across the world (EFMA 2000c, DOE 2000), and its derivativeCAN is particularly important fertiliser in Europe (Davis and Haglund 1999; Patyk1996). AN is produced by neutralising gaseous ammonia with aqueous nitric acid.The solution is evaporated and then formed into solid fertiliser by prilling orgranulation (EFMA 2000c). Before solidification, the solution may be mixed withdolomite or limestone to make CAN (EFMA 2000c). ‘Mean Nitrogen Fertiliser’ refersto a range of different fertiliser types used in Europe (see Table 5 for details).3.3.2. Emission Factors for AN, CAN and N Fertiliser ProductionTable 5 presents the GHG emissions from AN, CAN and Mean N FertiliserProduction. With the exception of West and Marland (2001) from the US, estimateswere based on plants in Western Europe. It should be noted here that the US estimateexcluded N 2 O emissions (West, pers. comm.), which are significant in total GHGemissions. Regardless of the omission of N 2 O, the US estimate is still relatively low.In all studies, the majority of these emissions are made up of CO 2 emissions fromammonia synthesis and N 2 O emissions from nitric acid production. Based onavailable information from each report, N 2 O emissions from nitric acid productionaccounted for an estimated 60-78% and 52-61% of total CO 2-e emissions fromAN/CAN and Mean N Fertiliser production respectively (Table 5). Emissions arisingfrom processing of intermediate products (i.e. ammonia and nitric acid) into final10

products (i.e. CAN, AN etc.) were of relatively minor importance (Patyk andReinhardt 1996; Davis and Haglund 1999).Data for emissions from transport of raw materials and intermediate products wereincluded in several studies. Patyk (1996), Patyk and Reinhardt (1996) and Davis andHaglund (1999) accounted for distances covered by truck, railway, inland navigationor boat, and calculated emissions based on published emissions factors. Neitherincluded transport for distribution of the final fertiliser product. The contribution oftransport to total emissions was minor (approximately 1% to 3%).The effect of the steam credits generated at various stages of the AN/CAN life cycleis difficult to interpret in terms of overall emissions, given the lack of transparency ofseveral reports. Data from Davis and Haglund (1999) suggest that the production ofsteam has only a minor impact on total GHG emissions (see ‘Steam’ in Table 5).Table 5: Greenhouse gas emission factors for Ammonium Nitrate (AN), Calcium AmmoniumNitrate (CAN) and Mean N Fertilisers.Product Country Composition g CO 2-e ReferenceN:P:K per kg N per kg Product CO 2:N 2O:CHTotal cSteam TransportANEuropean 35.0:0:0 7030.8 2460.8 -38.6 14.2 39.6:59.5:0.9 Davis and Haglund (1999)AverageANEuropean 33.5:0:0 6806.0 2280.0 - - - Kongshaug (1998)AverageANEurope 33.5:0:0 2985.1 1000.0 - - - Kongshaug (1998)Modern Tech.AN Netherlands 33.5:0:0 a 7108.7 2381.4 - - 38.5:60.9:0.5 Kramer (1999)ANUnited 33.5:0:0 a 6536.6 2189.8 - - 29.1:69.7:1.2 Elsayed (2003)KingdomAN Europe 33.5:0:0 a 6726.0 2253.2 - - 22.1:77.9;0.0 Kuesters and Jenssen(1998)CAN Sweden: 27.6:0:0 8467.1 2336.9 -25.3 25.9 38.3:60.9:0.8 Davis and Haglund (1999)LandskronaCAN Sweden: 27.6:0:0 9562.1 2639.2 -50.1 25.2 33.4:65.9:0.7 Davis and Haglund (1999)Koping 1CAN Sweden: 27.2:0:0 9562.9 2601.1 -48.8 24.8 33.4:65.9:0.7 Davis and Haglund (1999)Koping 2CAN Europe 26.5:0:0 a 7481.9 1982.7 -24.4 20.1 39.8:60.6:0.9 Davis and Haglund (1999)AverageCAN Europe 26.5:0:0 6867.9 1820.0 - - - Kongshaug (1998)AverageCAN Europe 26.5:0:0 3018.9 800.0 - - - Kongshaug (1998)Modern Tech.CAN Netherlands 27.9:0:0 6810.0 1900.0 - - 38.8:60.6:0.5 Kramer (1999)4Mean N Fert d Germany 28.6 7615.9 2178.1 - 59.2 36.6:61.4:1.9 Patyk and Reinhardt(1996)Mean N Fert d Germany 27.7 5339.9 1479.1 - 18.0 45.0:54.8:0.1 Patyk (1996)Mean N Fert d Germany 27.7 b 5644.6 1563.6 - - 46.9:52.9:0.2 Kaltschmitt and11

Reinhardt, 1997) fNitrogen Fert. US - 857.5 e - - - 100:0:0 West and Marland (2001)a) Composition of CAN and AN from Kongshaug (1998).b) Composition of Nitrogen Fertiliser from Patyk (1996)c) Transport not included in Total Emissions. Steam Credit was included in Total Emissions.d) Mean emissions and composition from Ammonia, Nitric Acid, CAN, Urea AN, MAP, DAP, ANP.e) This estimate is exclusive of N 2O emissions, see text.f) Kaltschmitt and Reinhardt (1997) cited in Elsayed et al. (2003).Note: Figures in italics are derived values, based on % N composition.3.4. Greenhouse Gas Emissions Factors for Urea and Urea-Ammonium NitrateProduction.3.4.1. Overview of Urea and Urea-Ammonium Nitrate ProductionUrea accounts for almost 50% of world nitrogen fertiliser production (UNEP 1996).The synthesis of urea is based on the combination of ammonia and carbon dioxide athigh pressure to form ammonium carbonate, which is subsequently dehydrated by theapplication of heat to form urea and water (EFMA 2000d). Liquid Urea-AmmoniumNitrate (UAN) is formed by mixing and cooling concentrated urea and ammoniumnitrate solutions (EFMA 2000d).3.4.2. Emission Factors for Urea and Urea-Ammonium Nitrate ProductionEmissions factors for Urea and UAN are given in Table 6. Emissions from ureaproduction are dominated by CO 2 emitted during ammonia synthesis. N 2 O emissionsaccount for a significant proportion of emissions from UAN production, becausenitric acid is an intermediate product in ammonium nitrate synthesis.Production of urea is usually linked to an ammonia plant, where by-product CO 2 fromammonia synthesis is used as a primary input in urea production. The discrepancybetween emissions factors for urea in Table 6 may be attributed to how this utilisationof by-product CO 2 is interpreted within the fertiliser life cycle. Because Kongshaug(1998) and Kuesters and Jenssen (1998) were not conducting full LCA studies, theyconsider that this consumption of CO 2 constitutes a net reduction in by-product CO 2emissions, hence the comparatively low emission factors. In contrast, Davis andHaglund (1999) did not include this net credit under the assumption that by-productCO 2 is only stored for a short time, and is eventually released upon application of ureafertilisers in the field. The IPCC (1996a, Section 2.13) recommends that “no accountshould be taken for intermediate binding of CO 2 in downstream manufacturingprocesses and products”.12

Table 6: Greenhouse Gas Emission Factors for Urea and Urea Ammonium Nitrate (UAN)Production.Produc Country Composition g CO 2-e ReferencetN:P:K per kg N per kg CO 2:N 2O:CH 4ProductUrea Europe Average 46:0:0 4018.9 1848.7 97.5:0.1:2.3 Davis and Haglund (1999)Urea Europe Average 46:0:0 1326.1 610.0 - Kongshaug (1998)Urea Europe: Modern46:0:0 913.0 420.0 - Kongshaug (1998)Tech.Urea Europe 46:0:0 a 1707.3 785.4 - Kuesters and Jenssen(1998)UAN Europe 32:0:0 a 3668.0 1173.8 36.6:63.4:0.0 Kuesters and Jenssen(1998)UAN Europe Average 32:0:0 5762.9 1844.1 59.1:39.5:1.4 Davis and Haglund (1998)UAN Europe Average 32:0:0 4093.8 1310.0 - Kongshaug (1998)UAN Europe Modern32:0:0 2000.0 640.0 - Kongshaug (1998)Tech.(a) Composition from Kongshaug (1998).Note: Figures in italics are derived values, based on % N composition.4. Greenhouse Gas Emission Factors for Phosphate FertilisersIn this section, published GHG emission factors for key phosphate fertilisers arepresented and discussed. Emissions factors for Single Superphosphate (SSP), TripleSuperphosphate (TSP), Diammonium Phosphate (DAP), Monoammonium Phosphate(MAP) and a ‘Mean Phosphate Fertiliser’ are provided. A discussion of theproduction process for phosphate fertilisers and the primary sources of emissions isincluded.4.1. Overview of Phosphate Fertiliser ProductionThe phosphate fertilisers discussed here are produced from various combinations ofphosphate rock, sulphuric acid, phosphoric acid and ammonia:• Single Superphosphate (SSP): phosphate rock and sulphuric acid,• Triple Superphosphate (TSP): phosphate rock and phosphoric acid,• Mono- and Di- Ammonium Phosphate (MAP/DAP): phosphoric acid andammonia.The majority of the world’s phosphate fertilisers are based on phosphoric acid(Kongshaug 1998). Fertiliser grade phosphoric acid is produced using the ‘wetprocess’, where sulphuric acid is reacted with naturally occurring phosphate rock13

(EFMA 2000f; DOE 2000). Apatite minerals comprise the primary phosphate rockinputs. Phosphoric acid may be produced via the dihydrate process or more energyefficient hemihydrate process. The US and European phosphoric acid manufacturerstypically use the dihydrate process (EFMA 2000f; DOE 2000).Sulphuric acid is required for the production of phosphoric acid. More sulphuric acidis produced than any other chemical in the world and the largest single user is thefertiliser industry (EFMA 2000e). Sulphuric acid is usually made using the contactmethod, an oxidation process based on the burning of elemental sulfur (brimstone)(DOE 2000). Elemental sulphur is generally sourced from the oxidation of hydrogensulphide from natural gas or crude oil, but can also come from sulphur mining(EFMA 2000e; DOE 2000). The sulphuric acid conversion process is highlyexothermic, often resulting in a net export of energy from the more efficient sulphuricacid plants.Ammonium phosphate fertilisers (MAP and DAP) are found in both solid and liquidforms and are produced by the reaction of phosphoric acid with anhydrous ammonia.Single superphosphate (SSP) is made by reacting ground phosphate rock with variousconcentrations of sulfuric acid and Triple superphosphate (TSP) is produced bycombining ground phosphate rock or limestone with low concentration phosphoricacid (DOE 2000).4.2. Greenhouse Gas Emission Factors for Phosphate FertilisersGHG emissions factors for phosphate fertilisers were available from six studies(Table 7). Most estimates were based on studies of European systems, with the soleexception being the study of West and Marland (2001) from the US. Emissionsestimates are dominated by CO 2 . Most of these emissions are related to theconsumption of fossil fuels as an energy source for the various production processesinvolved in phosphate fertiliser production.The net emission from phosphate fertiliser manufacture is largely determined by themethod of sulphuric acid production. The estimates in Table 7 all assumed that themajority of sulphur used is recovered from natural gas and fuel oil. Consequently, no14

emissions are linked to the sulphur raw material. The exothermic reactions involvedin the production of sulphuric acid may generate a net energy export. This may betranslated as an emissions credit (see Table 2) and has considerable bearing onemissions estimates. The effect of this emissions credit is well illustrated by thenegative accumulated emissions estimate for Modern European DAP/MAP andSSP/TSP plants that are equipped with technology capable of capitalising on thisenergy surplus (Table 7; ‘Europe Modern Tech’, Kongshaug 1998).Patyk (1996), Patyk and Reinhardt (1996) and Davis and Haglund (1999) eachprovided data on emissions from transport of raw materials and intermediate productsduring phosphate fertiliser manufacture. Transport comprised a considerableproportion of the emissions budget, making up about one-third of total emissions inPatyk (1996) and Patyk and Reinhardt (1996) and 20-25% of total emissions in Davisand Haglund (1999). Estimates took into account transport distances and appropriateemission factors for each mode of transport (i.e. sea, rail, road). The overseastransport of raw phosphate and P fertiliser was particularly important in the studies ofPatyk (1996) and Patyk and Reinhardt (1996).Insufficient information was provided to identify the source of the large discrepancybetween ‘European Average’ emissions estimates from Davis and Haglund (1999)and Kongshaug (1998).Table 7: Greenhouse Gas Emission Factors for Phosphate Fertilisers.Product Country Comp’n g CO 2-e ReferenceN:P:K:S dPer kg N per kg P 2O 5dper kg product CO 2:N 2O:CH 4Total aTransportSSP Europe Average 0:21:0:23 - 1051.8 220.9 70.4 93.3:1.1:5.7 Davis and Haglund(1999)SSP Europe Average 0:21:0:23 - 95.2 20.0 - - Kongshaug (1998)SSPEurope ModernTech.0:21:0:23 - -238.1 -50.0 - - Kongshaug (1998)TSP Europe Average 0:48:0:0 - 1083.5 520.1 170.5 96.7:0.5:2.8 Davis and Haglund(1999)TSP Europe Average 0:48:0:0 - 354.2 170.0 - - Kongshaug (1998)TSPEurope ModernTech.0:48:0:0 - -416.7 -200.0 - - Kongshaug (1998)N and P Compound FertiliserMAP Europe Average 11:52:0:0 6392.9 1352.4 703.2 238.1 97.9:0.3:1.9 Davis and Haglund(1999)MAP Europe Average 11:52:0:0 2818.2 596.2 310.0 - - Kongshaug (1998)MAPEurope ModernTech.11:52:0:0 -2454.5 -519.2 -270.0 - - Kongshaug (1998)15

DAP Europe Average 18:46:0:0 4812.0 1883.0 866.2 211.3 97.8:0.2:1.9 Davis and Haglund(1999)DAP Europe Average 18:46:0:0 2555.6 1000.0 460.0 - - Kongshaug (1998)DAPEurope ModernTech.18:46:0:0 -388.9 -152.2 -70.0 - - Kongshaug (1998)Mean PFert c Germany 0:32.2:0:0 - 817.3 263.2 116.1 93.8:1.0:5.2 Patyk and Reinhardt(1996)Mean P Germany 0:38.8:0:0 - 458.0 177.7 103.0 97.8:2.1:0.1 Patyk (1996)Fert cMean PFert c Germany 0:35.5:0:0 b - 700.0 248.5 100:0.0:0.0 Kaltschmitt andReinhardt (1996) eP Fertiliser US - - 165.1 - 100:0.0:0.0 West and Marland(2001)a) Total emissions do not include Transport.b) Assumed as average of Patyk (1996) and Patyk and Reinhardt (1996)c) Mean P Fertiliser includes Phosphoric acid, SSP, TSP, MAP, DAP and ANP.d) For West European products the grades are given in weight fraction of Nitrogen (N), phosphorus oxide (P 2O 5) and potassiumoxide (K 2O)e) Kaltschmitt and Reinhardt (1997) cited in Elsayed et al. (2003).Note: Figures in italics are derived values, based on % N or P 2O 5 composition.5. Greenhouse Gas Emission Factors for NPK Fertilisers5.1. Overview of NPK Fertiliser ProductionThere are several ways to produce multi-nutrient NPK fertilisers. In Europe, the twocommon routes are the nitrophosphate route and the mixed acid route (EFMA 2000g).An alternative method involves simply mixing dry fertilisers. The mixed acid routerequires phosphoric, sulphuric and nitric acids as raw materials. These acids aremixed and then neutralised with gaseous ammonia. Other compounds containingpotassium and magnesium are subsequently added. In the nitrophosphoric acid route,the first step involves reacting phosphate rock with an excess of nitric acid to producea mixture of nitric and phosphoric acid and calcium nitrate. The calcium nitrate isextracted, and the remaining solution is then neutralised with ammonia. Potassium(K) is added as potassium chloride (KCl) or Potassium sulphate (K 2 SO 4 ) salts.5.2. Greenhouse Gas Emission Factors for NPK FertilisersGHG emission factors were available from only two studies (Table 8). Estimates werebased on emissions data for Swedish NPK producers, average Europeanmanufacturers and European modern technology. The composition of these fertilisersand the processes used to produce them varied markedly. Most of the CO 2 emissionsoriginated from ammonia production because of the large consumption of fossil fuels16

(see Section 3). Almost 100% of total N 2 O emissions were released during theproduction of nitric acid; other processes such as the extraction of rock phosphate andproduction of sulphuric acid and phosphoric acid were not important (Kongshaug,1999).Table 8: Greenhouse Gas Emissions from NPK Fertilisers.Product Country Comp’n g CO 2-e ReferenceN:P:K b per kg N per kg P per kg K per kg product CO 2:N 2O:CH 4Total aT’sportNPK Sweden:Koping 17:04:13 9416.0 40018.2 12313.3 1600.7 29.9 40.2:59.0:0.8 Davis and Haglund(1999)NPK Sweden:Koping 20:05:04 9545.8 38183.2 47729.0 1909.2 29.9 37.9:61.4:0.8 Davis and Haglund(1999)NPK Sweden:Koping 21:03:10 7973.6 55815.0 16744.5 1674.5 30.2 38.3:60.9:0.8 Davis and Haglund(1999)NPK Sweden:Koping 21:04:07 8660.1 45465.3 25980.2 1818.6 31.3 39.6:59.7:0.8 Davis and Haglund(1999)NPK Sweden:Koping 24:04:05 8602.4 51614.2 41291.3 2064.6 32.1 38.3:61.0:0.8 Davis and Haglund(1999)N:P:K b per kg N per kg P 2O 5 per kg K 2O per kg Product CO 2:N 2O:CH 4Total aT’sportNPK (Acid) Europe Average 15:15:15 7495.6 7495.6 7495.6 1124.3 86.4 52.1:46.9:1.0 Davis and Haglund(1999)NPK (Nitro) Europe Average 15:15:15 7895.7 7895.7 7895.7 1184.4 39.5 50.8:48.2:1.1 Davis and Haglund(1999)NPK (Acid) Europe: Average 15:15:15 6466.7 6466.7 6466.7 970.0 Konshaug (1998)NPK (Nitro) Europe: Average 15:15:15 5533.3 5533.3 5533.3 830.0 Konshaug (1998)NPK (Mix) Europe: Average 15:15:15 2600.0 2600.0 2600.0 390.0 Konshaug (1998)NPK (Mix) Europe: Average 15:15:15 2266.7 2266.7 2266.7 340.0 Konshaug (1998)NPK (Acid) Europe: Mod 15:15:15 2133.3 2133.3 2133.3 320.0 Konshaug (1998)Tech.NPK (Nitro) Europe: Mod 15:15:15 2733.3 2733.3 2733.3 410.0 Konshaug (1998)Tech.NPK (Mix) Europe: Mod 15:15:15 400.0 400.0 400.0 60.0 Konshaug (1998)Tech.NPK (Mix) Europe: Mod 15:15:15 800.0 800.0 800.0 120.0 Konshaug (1998)Tech.a) Total emissions do not include transport.b) According to Swedish standard, the grades are given in weight fraction (%) of nitrate (N), phosphorus (P) and Potassium (K). In thecase of West European products the grades are given in weight fraction of Nitrogen (N), phosphorus oxide (P 2O 5) and potassiumoxide (K 2O)Note: Figures in italics are derived values, based on % N or P 2O 5 composition.17

6. SummaryThis report collates published greenhouse gas (GHG) emission factors associated withthe production of a range of nitrogen, phosphate and multi-nutrient fertilisers, for usein agricultural and forestry Life Cycle Assessments. Overall, there were fewpublished studies that present GHG emission factors for fertiliser production and, withthe exception of one study from the US, all were based on fertiliser manufacture inWestern Europe. Emission factors for the intermediate products Ammonia and NitricAcid were more widely available, but these are less useful in an agricultural LCAcontext.The major GHG emissions associated with nitrogen-containing fertiliser productionare carbon dioxide (CO 2 ) emitted when natural gas is combusted as part of ammoniasynthesis, and nitrous oxide (N 2 O) emitted during nitric acid production. ForPhosphate Fertilisers, GHG emissions were primarily CO 2 emitted during theconsumption of fossil fuels used in the various production processes. Transport of rawmaterials was an important contributor to phosphate fertiliser GHG emissions.For each fertiliser product, emissions factors varied markedly between studies. Thisvariation is due to differences in plant design and efficiency, emissions controlmechanisms and raw material inputs. Differences may also be attributed toassumptions made by the analysts during the calculation of the emission factors,particularly in relation to the interpretation of energy and emissions credits andtransport considerations. An understanding of these factors is required to make aninformed decision as to which emissions factor is most applicable for the particularfertiliser product being considered for a specific study.18

7. ReferencesDavis, J. and Haglund, C. 1999. Life Cycle Inventory (LCI) of Fertiliser Production.Fertiliser Products Used in Sweden and Western Europe. SIK-Report No. 654.Masters Thesis, Chalmers University of Technology.Elsayed, M. A., Matthews, R. and Mortimer, N. D. 2003. Carbon and EnergyBalances for a Range of Biofuels Options. Report by Resources Research Unit,Sheffield Hallam University for DTI Sustainable Energy Programmes.Engelstad, O. P. 1985. Fertilizer Technology and Use. Third Edition. Soil ScienceSociety of America, Inc. Madison, Wisconsin, USA.European Fertilizer Manufacturers’ Association (EFMA) 2000a. Production ofAmmonia. Booklet No. 1 of 8: Best Available Techniques for Pollution Control in theEuropean Fertilizer Industry. Ave. E van Nieuwenhuyse 4, B-1160 Brussels, Belgium.European Fertilizer Manufacturers’ Association (EFMA) 2000b. Production of NitricAcid. Booklet No. 2 of 8: Best Available Techniques for Pollution Control in theEuropean Fertilizer Industry. Ave. E van Nieuwenhuyse 4, B-1160 Brussels, Belgium.European Fertilizer Manufacturers’ Association (EFMA) 2000c. Production ofAmmonium Nitrate and Calcium Ammonium Nitrate. Booklet No. 6 of 8: BestAvailable Techniques for Pollution Control in the European Fertilizer Industry. Ave.E van Nieuwenhuyse 4, B-1160 Brussels, Belgium.European Fertilizer Manufacturers’ Association (EFMA) 2000d. Production of Ureaand Urea Ammonium Nitrate. Booklet No. 5 of 8: Best Available Techniques forPollution Control in the European Fertilizer Industry. Ave. E van Nieuwenhuyse 4, B-1160 Brussels, Belgium.European Fertilizer Manufacturers’ Association (EFMA) 2000e. Production ofSulphuric Acid. Booklet No. 3 of 8: Best Available Techniques for Pollution Controlin the European Fertilizer Industry. Ave. E van Nieuwenhuyse 4, B-1160 Brussels,Belgium.European Fertilizer Manufacturers’ Association (EFMA) 2000f. Production ofPhosphoric Acid. Booklet No. 4 of 8: Best Available Techniques for Pollution Controlin the European Fertilizer Industry. Ave. E van Nieuwenhuyse 4, B-1160 Brussels,Belgium.European Fertilizer Manufacturers’ Association (EFMA) 2000g. Production of NPKFertilisers by the Nitrophosphoric Acid Route. Booklet No. 7 of 8: Best AvailableTechniques for Pollution Control in the European Fertilizer Industry. Ave. E vanNieuwenhuyse 4, B-1160 Brussels, Belgium.EFMA 2000h. cited in “Calculating CO2 emissions from ammonia production”.http://www.ghgprotocol.org/standard/convert_tools_10_2_03/ammonia_guidance.v1.0.doc. Last Accessed March 29, 2004.19

Intergovernmental Panel on Climate Change (IPCC) 1996a. Revised IPCC Guidelinesfor National Greenhouse Gas Inventories: Reference Manual, London, U.K.Intergovernmental Panel on Climate Change (IPCC) 1996b. Revised IPCC Guidelinesfor National Greenhouse Gas Inventories: Workbook, Volume 2, London, U.K.Intergovernmental Panel on Climate Change (IPCC) 2000 Good Practice Guidanceand Uncertainty Management in National Greenhouse Gas Inventories: N 2 OEmissions from Adiphic Acid and Nitric Acid Production. London, U.K.Kongshaug, G. 1998. Energy Consumption and Greenhouse Gas Emissions inFertilizer Production. IFA Technical Conference, Marrakech, Morocco, 28September-1 October, 1998, 18pp.Kramer, K.J., Moll, H.C. and Nonhebel, S. 1999. Total Greenhouse Gas EmissionsRelated to the Dutch Crop Production System. Agriculture, Ecosystems andEnvironment, 72, 9-16.Kuesters, J. and Jenssen, T. 1998 Selecting the Right Fertilizer from anEnvironmental Life Cycle Perspective. IFA Technical Conference, Marrakech,Morocco, 28 September-1 October, 1998, 7pp.Patyk, A. 1996. Balance of Energy Consumption and Emissions of FertilizerProduction and Supply. Reprints from the International Conference of Life CycleAssessment in Agriculture, Food and Non-Food Agro-Industry and Forestry:Achievements and Prospects, Brussels, Belgium, 4-5 April 1996.Patyk, A. and Reinhardt, G. A. 1996. Energy and Material Flow Analysis of FertilizerProduction and Supply. In SETAC-Europe (Society of Environmental Toxicology andChemistry) (ed.): Presentation Summaries of the 4th Symposium for Case Studies, 3December 1996 in Brussels, pp. 73-87, Brussels 1996West, T. O. and Marland, G. 2001. A Synthesis of Carbon Sequestration, CarbonEmissions and Net Carbon Flux in Agriculture: Comparing Tillage Practices in theUnited States. Agriculture, Ecosystems and Environment 1812, 1-16.US Department of Energy (DOE) 2000. Agricultural Chemicals: Fertilisers. Chapter 5in Energy and Environmental Profile of the U.S. Chemical Industry.US Environmental Protection Agency (EPA) 1993 Synthetic Ammonia, AP-42, 5thed., Volume 1, Chapter 8, Environmental Protection Agency.UNEP 1996. Mineral Fertiliser Production and the Environment, A Guide toReducing the Environmental Impact of Fertiliser Production. Technical report No. 26.United Nations Environment Programme Industry and the Environment, Paris.20